Integrate zhouwang/hiddenlayername into master

MikalaiDrabovich · Sep 23, 2016 · 667c9e5 · 667c9e5
2 parents 689b473 + a6462c3
commit 667c9e5
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 16 deletions.
diff --git a/Examples/Evaluation/CSEvalClient/Program.cs b/Examples/Evaluation/CSEvalClient/Program.cs
@@ -174,8 +174,8 @@ private static void EvaluateModelMultipleLayers()
                 using (var model = new IEvaluateModelManagedF())
                 {
                     // Desired output layers
-                    const string hiddenLayerName = "h1.z";
-                    const string outputLayerName = "ol.z";
+                    const string hiddenLayerName = "out.h1";
+                    const string outputLayerName = "out.z";
 
                     // Load model
                     string modelFilePath = Path.Combine(Environment.CurrentDirectory, @"..\Output\Models\01_OneHidden");

diff --git a/Examples/Image/MNIST/Config/01_OneHidden.cntk b/Examples/Image/MNIST/Config/01_OneHidden.cntk
@@ -16,36 +16,37 @@ stderr = "$outputDir$/01_OneHidden_bs_out"
 # TRAINING CONFIG
 trainNetwork = {
     action = "train"
-    
+
     BrainScriptNetworkBuilder = {
         imageShape = 28:28:1                        # image dimensions, 1 channel only
         labelDim = 10                               # number of distinct labels
-        featScale = 1/256
-        Scale{f} = x => Constant(f) .* x
-
-        model = Sequential (
-            Scale {featScale} :
-            DenseLayer {200, activation=Sigmoid} :
-            LinearLayer {labelDim}
-        )
-
+        featScale = 1/256     
+
+        # This model returns multiple nodes as a record, which
+        # can be accessed using .x syntax.
+        model(x) = {
+            s1 = x * featScale
+            h1 = DenseLayer {200, activation=Sigmoid} (s1)
+            z = LinearLayer {labelDim} (h1)
+        }
+
         # inputs
         features = Input {imageShape}
         labels = Input (labelDim)
 
         # apply model to features
-        ol = model (features)
+        out = model (features)
 
         # loss and error computation
-        ce   = CrossEntropyWithSoftmax (labels, ol)
-        errs = ClassificationError (labels, ol)
+        ce   = CrossEntropyWithSoftmax (labels, out.z)
+        errs = ClassificationError (labels, out.z)
 
         # declare special nodes
         featureNodes    = (features)
         labelNodes      = (labels)
         criterionNodes  = (ce)
         evaluationNodes = (errs)
-        outputNodes     = (ol)
+        outputNodes     = (out.z)
     }
 
     SGD = {